HIV integrase (IN) is a viral enzyme required for integration of virus DNA into the host chromosome. This type of integration is highly specific for retroviruses, and HIV-1 IN is therefore a target for antiviral therapy. Disruption of integrase function is also desirable in design of live attenuated HIV vaccines. We have expressed the IN gene in E. coli as a fusion protein. The cloned IN is reactive with both HIV-1 and HIV-2 positiv patient sera in ELISA, while rabbit antisera to the recombinant protein are reactive only with HIV-1 IN but not HIV-2 IN by Western blot. Hybridomas have been prepared using the IN expressing clone and the MAbs are reactive with HIV-1 but not HIV-2 by Western blot. An invention report has been file and a manuscript is in preparation. Additional clones expressing the N- and C-terminal halves of IN have been constructed and the proteins they express examined for ability to bind DNA using a Southwestern blotting procedure. The complete IN molecule as well as the C-terminal protein bind DNA; the N- terminal portion exhibits no binding activity, suggesting that the C- terminal region contains the DNA binding site. Purified full-length recombinant IN exhibits activity in a specific in vitro assay for enzyme activity involving cleavage of 2 bp from an oligonucleotide corresponding t the HIV-1 LTR. To more precisely localize the DNA binding site within the C terminal half of the molecule, we have constructed a series of additional subclones and analyzed the deleted proteins for DNA binding. This has enabled us to map the DNA binding site to amino acids 180-248. A manuscript is in press in Nucl. Acids Res. and preparation of additional deletions is in progress. We are also collaborating with David Davies at the NIH and providing him with large quantities of IN MAb for generation of Fab fragments for X-ray crystallography of IN, which may eventually facilitate design of enzyme inhibitors.